Action Detection On Ucf Sports
Metrics
Frame-mAP 0.5
Video-mAP 0.2
Video-mAP 0.5
Results
Performance results of various models on this benchmark
Model Name | Frame-mAP 0.5 | Video-mAP 0.2 | Video-mAP 0.5 | Paper Title | Repository |
|---|---|---|---|---|---|
| MR-TS R-CNN | 84.52 | 94.83 | 94.67 | Multi-region two-stream R-CNN for action detection | - |
| TS R-CNN | 82.30 | 94.82 | 94.82 | Multi-region two-stream R-CNN for action detection | - |
| Action Tubes | 68.1 | - | 75.8 | Finding Action Tubes | |
| Two-in-one Two Stream | - | - | 96.52 | Dance with Flow: Two-in-One Stream Action Detection | |
| Two-in-one | - | - | 92.74 | Dance with Flow: Two-in-One Stream Action Detection | |
| T-CNN | 86.7 | - | - | Tube Convolutional Neural Network (T-CNN) for Action Detection in Videos | |
| DTS | - | 94.3 | 93.8 | Finding Action Tubes with a Sparse-to-Dense Framework | - |
0 of 7 row(s) selected.